Thread: [pgadmin-hackers] User Research Findings into PgAdmin

Hi

On Tue, Jan 24, 2017 at 9:42 PM, Raffi Holzer <rholzer@pivotal.io> wrote:
> Hi,
>  As part of a project to enable pgAdmin4 to work with Greenplum databases,
> we conducted user research into pgAdmin. We wanted to share those insights
> and potentially develop against them but first wanted to get feedback from
> the pgAdmin group. To that end we created the video below that summarizes
> our findings. Let us know what you think.
>
> https://youtu.be/BylcblYw8sU

That's very helpful, thank you. Can you please summarise your findings
in an email to the list so we can work with it within our normal
process?

I will note that whilst I agree some of the points raised need further
work in pgAdmin 4, some of them make little sense - most notably
Syntax Highlighting which we've had for ~17 years, and works
automatically, thus not requiring any kind of user discovery beyond
typing some SQL into the query tool. Auto-complete is another example
of a feature we've had for years, that is documented. I agree though
that there is scope for better discovery mechanisms there.

Some of the other comments highlighted made little to no sense to me,
for example the comment on column names @ ~3:03 and the comment on the
bottom left @ ~3:22 (this is a great example of why email is better
than a video :-) ).

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Hi Dave, 

 We can definitely send a summary via email. Just to clarify those points you brought up. Syntax highlighting definitely does exist and is totally discoverable but people still seemed to miss it. Our best guess as to why so many users miss it is because of the relatively small contrast in colors (Navy vs black). Regarding autocomplete, despite it being documented, almost none of the long time customers we spoke with knew it existed which suggests they're not reading the documentation very closely or at all.

Regarding the comments you mentioned, the one regarding column names was talking about preserving the column name when moving data to Excel. And I'll admit, the other comment is pretty unclear out of context. We probably should not have included it as is.

A summary of the findings will be forthcoming.

On Wed, Jan 25, 2017 at 4:25 AM, Dave Page <dpage@pgadmin.org> wrote:

Hi

On Tue, Jan 24, 2017 at 9:42 PM, Raffi Holzer <rholzer@pivotal.io> wrote:
> Hi,
>  As part of a project to enable pgAdmin4 to work with Greenplum databases,
> we conducted user research into pgAdmin. We wanted to share those insights
> and potentially develop against them but first wanted to get feedback from
> the pgAdmin group. To that end we created the video below that summarizes
> our findings. Let us know what you think.
>
> https://youtu.be/BylcblYw8sU

That's very helpful, thank you. Can you please summarise your findings
in an email to the list so we can work with it within our normal
process?

I will note that whilst I agree some of the points raised need further
work in pgAdmin 4, some of them make little sense - most notably
Syntax Highlighting which we've had for ~17 years, and works
automatically, thus not requiring any kind of user discovery beyond
typing some SQL into the query tool. Auto-complete is another example
of a feature we've had for years, that is documented. I agree though
that there is scope for better discovery mechanisms there.

Some of the other comments highlighted made little to no sense to me,
for example the comment on column names @ ~3:03 and the comment on the
bottom left @ ~3:22 (this is a great example of why email is better
than a video :-) ).

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

--

Raffi Holzer

Product Manager

Pivotal Labs

My main concern is making the right decisions on what to change, and
focussing efforts in the areas where we gain most benefit. Of course,
the resources to do the work are also critical - the current team
already have a good years worth of work in various additional features
and improvements.

Absolutely our concern as well which is why we don't want to work in a vacuum. 

-- Rob

 

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Hi

On Wed, Jan 25, 2017 at 8:02 PM, Raffi Holzer <rholzer@pivotal.io> wrote:

As a summary of our findings in writing: 

Below is a table of user needs we discovered and solutions we or the customers suggested. Some of the "missing" features people complained about are not missing at all (such as auto-complete or syntax highlighting) but the fact that people didn't know about them points to a discoverability issue we believe we can address through design.

 

Needs	Suggested Solutions
Users need a reliable way to save their queries frequently	Autosave, Persisting & Complete History, Persisting Un-run queries after disconnect

Please add feature requests for:

- Auto-saving of queries (I believe this would implement persisting un-run queries as well).

- Saving of previous queries (implemented in pgAdmin III as a combo box at the top of the editor)

 

Users need a more efficient way to type

Auto-discoverable Autocomplete, Customize shortcuts, more visible syntax highlighting

Please add feature requests for:

- Hint mechanism, to display balloon hints that can be triggered during certain tasks to inform the user of functionality they might want to use

- Customisable shortcuts (a panel under Preferences -> [Tool Name] to allow customisation of shortcuts in each tool)

- Customisable syntax highlight colours

 

Users need a process to share query output with others

Copy and Paste into Excel with proper formatting, Export button to MS XML

Please add a feature request for:

- Allow customisation of the data copy formatting as found in pgAdmin III. The defaults should be changed to not quote strings and use a tab as the delimiter to allow pasting into Excel/Numbers. Note that tabs cannot be represented in TSV data, so we need to deal with that and warn the user. Pasting back into pgAdmin also needs to support format changes.

- Add a drop-down menu to the Export button that the user can use to select data format (the default remaining CSV). This can include XML, JSON and the user-defined copy format (per above).

 

Users need a way to quickly input content into pgAdmin

Edit within graphic table

I don't understand that - you can already add/edit/delete data in a grid.

 

Users need query output to be stored to reuse results

Results stored in history

Typically users would use multiple query tool windows for that. Saving the results in the history doesn't seem feasible - they can easily run into 10s of MB or more.

 

Users need an easier way to find saved queries

Workspace, Search

This would need much more definition. In pgAdmin III we had Macros that allowed you to insert standard queries you had saved in response to a shortcut key, and Favourites that allowed you to store a library of favourite queries and load or inject them into the editor. Plus of course, you can save queries to the file system. What would help here?

Oh, and before you ask: https://redmine.postgresql.org/projects/pgadmin4/issues/new :-)

Thanks.

 

On Wed, Jan 25, 2017 at 10:08 AM, Robert Eckhardt <reckhardt@pivotal.io> wrote:

My main concern is making the right decisions on what to change, and
focussing efforts in the areas where we gain most benefit. Of course,
the resources to do the work are also critical - the current team
already have a good years worth of work in various additional features
and improvements.

Absolutely our concern as well which is why we don't want to work in a vacuum. 

-- Rob

 

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

--

Raffi Holzer

Product Manager

Pivotal Labs

--

Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company